{ # X O } Space

     In this section, the Tic-Tac-Toe example is viewed as a combinatoric system defined over the alphabet # for blank, and X and O to refer to the moves. A state of the Tic-Tac-Toe game is then represented as a string composed of these elements. For example the string ####X#O## corresponds to the 3x3 Matrix case where the X is in the center cell of the 3x3 and the O is in the lower left cell. XX###O is also a string composed from this alphabet but it does not correspond to any TicTacToe state since the string's length is 6 and all of the TicTacToe strings are of length 9.

     The figure to the right depicts some of the possible strings that can be composed from this vocabulary. Note that the number of elements of the vocabulary is 3. The various ellipses depict all of the possible strings formed from this vocabulary of length 0 through 10. The number of strings in each of these sets is also shown. This value is simply 3, the number of elements in the vocabulary, raised to the power equal to the length of the string. For example, there are 3 to the 9 or 19,683 strings of length 9.

     The continuation dots that appear below the ellipse of length 10 is intended to remind us that we could continue to generate these sets for any length n as long as n is finite. The capital Greek letter Sigma followed by an * is usually used to refer to the set of all sets of strings up to length n. This is read as Sigma star. The large ellipse is intended to represent this set in the figure to the right. Only the sets 0 to 10 are explicitly represented. Note that they are each subsets of Sigma star.

     Clicking on any of these subsets (except for Sigma 10) will take you to a page that explicitly lists all of the strings that are elements of this set. Note that Sigma 9 contains 19,683 such strings and if you click on it it will take quite a while to load. These sets have been explicitly made available to you and the number of elements in each of these subsets computed so that you can begin to appreciate the size of unconstrained combinatoric spaces even when the vocabulary or elements of the system are small.

     Consider this construction in light of the levels hypothesis. Consider this unconstrained space as the physical level. That is there is no physical law that precludes writing down any sequence from this vocabulary of any arbitrary length. However, the game of Tic-Tac-Toe is only realized on strings that are exactly of length 9. This subset is shown in green in the figure above and is still quite large. But note that this is a "law" of Tic-Tac-Toe at the "symbol level," and not a law that holds at the "physical level." A subset of this set of strings of length 9 is represented by the blue ellipse. This subset has only 509 strings. This reduction was achieved by using additional "laws" of Tic-Tac-Toe to compute this subset. This subset contains only strings that have either the same number of  'Xs' and 'Os'; or exactly one more 'X' than the number of 'Os'. These laws reflect that X always begins and the players must alternate their moves.

     But a Tic-Tac-Toe game is a sequence of these strings where the sequence represents a particular sequence of moves. The next figure on the right helps to illustrate this idea. A sequence can be obtained by taking the cross-products of sigma star with itself. For example, the 4-tuple of strings (X#X, X###, O, XX) is an element that is obtained by a crossing sigma star with itself 4 times.

     The resulting spaces of possible sequences is, of course, really enormous. Illustrated to the right are subsets that are obtained by crossing sigma 9 with itself up to 9 times. Syntactically correct Tic-Tac-Toe games are a subset of the subset represented in green. Note that this set is itself enormous. It contains 19,863 to the 9 elements or 443,426,488,243,037,806,754,782,544,259,268,477,000 elements. If we consider only the 509 legal elements, this cross product is equal to 509 to the 9 which is still a rather large number of elements; namely 2,293,295,617,071,746,258,042,880 elements.

     However, if we take into account the laws of Tic-Tac-Toe that govern the transitions from one state to the next, then the number of possible sequences is less than 9! or 362,880.

     These enormous reductions in the sequences that are actually observed can not be explained by "physical laws", but they are explained by assuming that there are laws that govern the syntax or computation of these sequences.

     Finally, note that laws also govern a semantic level. One such law is: (a) make a winning move whenever and as soon as possible. Another is that if (a) is not possible, then block any imminent winning move by your opponent whenever possible. Note, nothing in the syntax of the game requires or leads to these laws. One can violate these rules and still be said to be "playing Tic-Tac-Toe." It is the semantics of competition that is represented in these laws. Again, the claim of the levels hypothesis is that if laws can't be found at the physical or syntactic level to account for observed constraints then a semantic level may be an appropriate level at which to describe the phenomena.

Levels Hypothesis | Introduction - Table of Contents

 © Charles F. Schmidt